Google Ads split-testing is ideally suited for determining which ads deliver the greatest return on investment for your e-commerce store.
Basically, by showing one ad to your audience half the time and a different ad the other half, you can better determine which option delivers the best click-through rates and cost-per-click.
Continually comparing and fine-tuning your ad campaigns will allow you to put the right ads in front of your audience, while also reducing ad spending.
With smarter Google Ads split-testing, you will be able to maximize the return on your ads like never before.
But not all split-testing is created equal.
In this post, we will break down seven split-testing techniques used by some of the most successful e-commerce marketers.
Google Ads split-testing strategy #1: Start With Boosted Organic Social Media Posts
Google Ads split-testing strategy #2: Focus on Buyer Basics
Google Ads split-testing strategy #3: Use Power Words
Google Ads split-testing strategy #4: Break Away from Conventional Wisdom
Google Ads split-testing strategy #5: Use Multiple Landing Pages
Google Ads split-testing strategy #6: Test Everything (Details Matter)
Google Ads split-testing strategy #7: Use Google’s Ad Variations
Ready to start?
Google Ads doesn’t need to be the first place you go to split-test new offers on your e-commerce products. One technique that I’ve found surprisingly effective is using organic social media posts to conduct early split tests.
The way this works is that when you come up with a new idea for a test variant, you create an organic social media post that is then boosted across your branded channels with a small promotion budget.
Facebook and other social media platforms provide plenty of their own data to help us check the click-through-rate so you can see how much interest actually exists for the offer among your current fans.
For example, KlientBoost not only offers full-blown PPC agency services, but they also sell a PPC/CRO academy for digital marketers. Back when they were more actively marketing this product, they used smoke testing to help identify the best way to generate registrations/students for their academy.
And what they found proves how important smoke testing can be:
Clearly, they were going about lead generation for registrations the entirely wrong way. People were more interested in taking home learning materials about the academy before scheduling a demo.
In fact, they were more than 4X as interested…
If there’s enough interest in the offer on social media (which safe to say there was), you then create a full-fledged page variant and begin the actual split testing with the landing page offer in Google Ads (where the real money is).
This saves you time and resources by ensuring that you don’t invest your marketing budget on a landing page and PPC campaign for an offer that doesn’t hold much interest with your target audience.
Many PPC campaigns start by focusing on demographics and psychographics. While this can certainly help when selecting targeting parameters for your campaign, it shouldn’t be the only thing that guides you as you write your PPC copy.
As Search Engine Watch’s Howard Jacobson explains:
PPC ads need to focus on “the situational who dimension, [which] relates to the search at hand. Someone searching for [motorcycle pants] might be getting their first motorcycle and have no idea what type of fit or material is right for them. They might be a seasoned street rider looking to take up motocross. They might be the bike owner or the passenger.”
By honing in on “the situational who,” you can then better determine the what, why, and how that goes into why a potential customer is typing those search terms into their browser in the first place.
Understanding these buyer basics will allow you to find the right voice and messaging by truly getting inside your customers’ heads — something that demographics and psychographics alone can’t accomplish. You’ll understand what they want, why they want it and how you can deliver it with your store’s products.
The most effective words — whether in a PPC campaign or a magazine ad — evoke emotion in the reader. These words are designed to inspire the customer to take immediate action to learn more about the product or make a purchase.
These are trigger words, also sometimes referred to as power words.
While marketing research has long proven that using the word “you” can have an impressive influence on buyers, this is far from the only power word that can evoke an immediate response.
A 2014 analysis of AdWords campaigns found that the most popular trigger words included:
Interestingly, all of these trigger words ranked higher than expected advertising lingo like “buy.”
Just be sure that your trigger words evoke the intended emotional response …
While “free” ranked as the most popular adjective, it won’t necessarily help you land paying customers, especially if you don’t actually have a free offer available.
Choose relevant power words so that customers don’t get disillusioned by your store content after they click on your ad.
There are countless articles espousing the “best practices” for PPC and other marketing techniques. While some PPC lessons (such as considering the lifetime value of your customers or having a clearly defined goal for your campaign) are always applicable, sometimes, breaking away from conventional wisdom can sometimes lead to better results.
A case study from HotJar is the perfect example of this. This software company, leaders in the heatmap analytics space, attempted to increase landing page registrations by following the oft-cited best practice of reducing the number of fields on online forms.
The number of registrations from anonymous visitors to signups skyrocketed… [However,] the remodeled form didn’t bring more new customers.
It only increased the amount of signups, but those signups were less qualified and less inclined to become customers.”
Louis Grenier, Content Lead at Hotjar
The same principle can just as easily apply to your store’s PPC campaigns.
When crafting running Google Ads split-testing, don’t be afraid to try something that goes against conventional PPC best practices.
For example, you could focus on product features rather than benefits in your ad copy, or use short keywords instead of long-tail keywords.
Each business, audience, and campaign is different. You never know when something unexpected might deliver stronger results than conventional marketing wisdom.
The ads themselves aren’t the only thing that you should split test. Sending users to different landing pages after they click on your site can also make a significant difference.
This gives you the opportunity to test different elements of your landing pages to determine which variants work best at converting customers after they click on an ad.
Even something as seemingly minor as adjusting the color of your call to action button or changing the information fields in a sign-up form can affect sales and leads.
In a case study from BigCommerce, Brandon Chatham, Founder and CEO of NatoMounts, noted that constantly performing A/B tests to optimize his funnels (for entire websites, not just landing pages) played a crucial role in streamlining the checkout process.
Consistently testing mobile optimizations and different payment integrations allowed him to get a new visitor to his landing pages to a completed checkout in as little as 43 seconds, resulting in over $20 million in revenue in a 6-year period.
Chatham achieved these results by focusing his website optimization on his largely mobile-centric audience.
For your own landing page tests, consider your audience’s preferred device type, interests and more. Don’t be afraid to go big with your adjustments, testing factors like the layout, product descriptions, images and more.
As you A/B test your landing pages, you will be better equipped to evaluate what resonates with your audience so you can improve engagement, conversions, and sales.
It’s all too easy to fall into the trap of testing wildly different ads against each other.
While split testing significantly different PPC copy can give you great insight into the type of content that best appeals to your target audience, sometimes, a surprisingly small change can yield just as big of a win.
Two PPC ads could use the same headline and URL, but change only a few words in the description, and one will perform significantly better than the other.
Is “learn more” a better call to action than “buy now?” What happens if you switch two of the lines within your text description, or take out a few commas? Incredibly, such changes can make a difference.
These small changes don’t have to be isolated to ad headlines either. Diving into the nitty-gritty of your account can sometimes help identify small adjustments you can make that can yield some pretty big wins.
For example, with one of our clients, we were doing some deep-dive analysis when we noticed a single change could be made within their bid strategy.
All we did was to switch from manual CPC to target CPA. And just look at the results;
With a 9% increase in conversions, an 11% increase in revenue, and a whopping 57% boost in ROAS, I think you can count that as a big win.
It just goes to prove that when it comes to CRO it’s not the time you put in, it’s what you put into the time.
Don’t be afraid to test seemingly insignificant facets of your ad. After all, increasing conversion rates even a fraction of a percent will result in more sales and revenue for your store.
Creating and testing a seemingly endless line of ad variations on your own can feel overwhelming, especially when you have other marketing responsibilities on your plate. Thankfully, with its aptly named ad variations system, Google makes it easy to create and test ads.
This system allows you to test variables on as much or as little of the ad as you want.
You could run variations on just the headline, just the description, or all facets of the ad. Filter options give you even more customization by helping you find the ads you wish to change.
Variation options include swapping your ad’s Headline 1 and Headline 2, finding and replacing a specific word or phrase from all ads in a campaign or updating text and URL options on an ad-by-ad basis.
Once the variation is ready, you can assign it a percentage of your campaign budget, and then wait to see the actual numbers come in.
After you’ve received your data, you can remove or pause the original ads and replace them with the new variation, or create new ads using the variation while still keeping the original ads in your PPC rotation. If the variation flopped, you can simply delete it.
But don’t just run one variation! You never know what kind of changes will yield the biggest impact.
The more variations you run, the easier it will be to find content that delivers the most for your metrics.
And Now Get To Testing!
PPC advertising remains one of the most powerful tools for e-commerce stores trying to build their customer base — but only if you create the right messaging with your Google Ads.
By split-testing your Google Ads, you will have greater confidence as you invest in your campaigns thanks to the actionable data you have discovered.
As you use Google Ads split-testing to learn more about your target audience, you will also be better positioned to craft even more successful ads in the future.
How has split-testing helped you improve your PPC campaign results? Have you tried any of these tips yourself? Leave a comment to share your story!
Johnathan Dane is an international speaker and the founder of KlientBoost, a no-nonsense, creative, kick-ass Google Ads and landing page agency that hustles for results and ROI.
If you think this article’s good, you should see what he’s writing on their PPC and CRO blog.
You can get in touch with him on Twitter, Facebook, Instagram, and LinkedIn.